Pangenome-spanning epistasis and coselection analysis via de Bruijn graphs.

Studies of bacterial adaptation and evolution are hampered by the difficulty of measuring traits such as virulence, drug resistance, and transmissibility in large populations. In contrast, it is now feasible to obtain high-quality complete assemblies of many bacterial genomes thanks to scalable high-accuracy long-read sequencing technologies. To exploit this opportunity, we introduce a phenotype- and alignment-free method for discovering coselected and epistatically interacting genomic variation from genome assemblies covering both core and accessory parts of genomes. Our approach uses a compact colored de Bruijn graph to approximate the intragenome distances between pairs of loci for a collection of bacterial genomes to account for the impacts of linkage disequilibrium (LD). We demonstrate the versatility of our approach to efficiently identify associations between loci linked with drug resistance and adaptation to the hospital niche in the major human bacterial pathogens Streptococcus pneumoniae and Enterococcus faecalis.

Robust analysis of prokaryotic pangenome gene gain and loss rates with Panstripe.

Horizontal gene transfer (HGT) plays a critical role in the evolution and diversification of many microbial species. The resulting dynamics of gene gain and loss can have important implications for the development of antibiotic resistance and the design of vaccine and drug interventions. Methods for the analysis of gene presence/absence patterns typically do not account for errors introduced in the automated annotation and clustering of gene sequences. In particular, methods adapted from ecological studies, including the pangenome gene accumulation curve, can be misleading as they may reflect the underlying diversity in the temporal sampling of genomes rather than a difference in the dynamics of HGT. Here, we introduce Panstripe, a method based on generalized linear regression that is robust to population structure, sampling bias, and errors in the predicted presence/absence of genes. We show using simulations that Panstripe can effectively identify differences in the rate and number of genes involved in HGT events, and illustrate its capability by analyzing several diverse bacterial genome data sets representing major human pathogens.

Fast and flexible bacterial genomic epidemiology with PopPUNK.

The routine use of genomics for disease surveillance provides the opportunity for high-resolution bacterial epidemiology. Current whole-genome clustering and multilocus typing approaches do not fully exploit core and accessory genomic variation, and they cannot both automatically identify, and subsequently expand, clusters of significantly similar isolates in large data sets spanning entire species. Here, we describe PopPUNK (Population Partitioning Using Nucleotide K -mers), a software implementing scalable and expandable annotation- and alignment-free methods for population analysis and clustering. Variable-length k-mer comparisons are used to distinguish isolates' divergence in shared sequence and gene content, which we demonstrate to be accurate over multiple orders of magnitude using data from both simulations and genomic collections representing 10 taxonomically widespread species. Connections between closely related isolates of the same strain are robustly identified, despite interspecies variation in the pairwise distance distributions that reflects species' diverse evolutionary patterns. PopPUNK can process 103-104 genomes in a single batch, with minimal memory use and runtimes up to 200-fold faster than existing model-based methods. Clusters of strains remain consistent as new batches of genomes are added, which is achieved without needing to reanalyze all genomes de novo. This facilitates real-time surveillance with consistent cluster naming between studies and allows for outbreak detection using hundreds of genomes in minutes. Interactive visualization and online publication is streamlined through the automatic output of results to multiple platforms. PopPUNK has been designed as a flexible platform that addresses important issues with currently used whole-genome clustering and typing methods, and has potential uses across bacterial genetics and public health research.

Early Signals of Vaccine-driven Perturbation Seen in Pneumococcal Carriage Population Genomic Data.

BACKGROUND: Pneumococcal conjugate vaccines (PCVs) have reduced pneumococcal diseases globally. Pneumococcal genomic surveys elucidate PCV effects on population structure but are rarely conducted in low-income settings despite the high disease burden. METHODS: We undertook whole-genome sequencing (WGS) of 660 pneumococcal isolates collected through surveys from healthy carriers 2 years from 13-valent PCV (PCV13) introduction and 1 year after rollout in northern Malawi. We investigated changes in population structure, within-lineage serotype dynamics, serotype diversity, and frequency of antibiotic resistance (ABR) and accessory genes. RESULTS: In children <5 years of age, frequency and diversity of vaccine serotypes (VTs) decreased significantly post-PCV, but no significant changes occurred in persons ≥5 years of age. Clearance of VT serotypes was consistent across different genetic backgrounds (lineages). There was an increase of nonvaccine serotypes (NVTs)-namely 7C, 15B/C, and 23A-in children <5 years of age, but 28F increased in both age groups. While carriage rates have been recently shown to remain stable post-PCV due to replacement serotypes, there was no change in diversity of NVTs. Additionally, frequency of intermediate-penicillin-resistant lineages decreased post-PCV. Although frequency of ABR genes remained stable, other accessory genes, especially those associated with mobile genetic element and bacteriocins, showed changes in frequency post-PCV. CONCLUSIONS: We demonstrate evidence of significant population restructuring post-PCV driven by decreasing frequency of vaccine serotypes and increasing frequency of few NVTs mainly in children under 5. Continued surveillance with WGS remains crucial to fully understand dynamics of the residual VTs and replacement NVT serotypes post-PCV.

A mosaic tetracycline resistance gene tet(S/M) detected in an MDR pneumococcal CC230 lineage that underwent capsular switching in South Africa.

OBJECTIVES: We reported tet(S/M) in Streptococcus pneumoniae and investigated its temporal spread in relation to nationwide clinical interventions. METHODS: We whole-genome sequenced 12 254 pneumococcal isolates from 29 countries on an Illumina HiSeq sequencer. Serotype, multilocus ST and antibiotic resistance were inferred from genomes. An SNP tree was built using Gubbins. Temporal spread was reconstructed using a birth-death model. RESULTS: We identified tet(S/M) in 131 pneumococcal isolates and none carried other known tet genes. Tetracycline susceptibility testing results were available for 121 tet(S/M)-positive isolates and all were resistant. A majority (74%) of tet(S/M)-positive isolates were from South Africa and caused invasive diseases among young children (59% HIV positive, where HIV status was available). All but two tet(S/M)-positive isolates belonged to clonal complex (CC) 230. A global phylogeny of CC230 (n=389) revealed that tet(S/M)-positive isolates formed a sublineage predicted to exhibit resistance to penicillin, co-trimoxazole, erythromycin and tetracycline. The birth-death model detected an unrecognized outbreak of this sublineage in South Africa between 2000 and 2004 with expected secondary infections (effective reproductive number, R) of ∼2.5. R declined to ∼1.0 in 2005 and <1.0 in 2012. The declining epidemic could be related to improved access to ART in 2004 and introduction of pneumococcal conjugate vaccine (PCV) in 2009. Capsular switching from vaccine serotype 14 to non-vaccine serotype 23A was observed within the sublineage. CONCLUSIONS: The prevalence of tet(S/M) in pneumococci was low and its dissemination was due to an unrecognized outbreak of CC230 in South Africa. Capsular switching in this MDR sublineage highlighted its potential to continue to cause disease in the post-PCV13 era.

Global emergence and population dynamics of divergent serotype 3 CC180 pneumococci.

Streptococcus pneumoniae serotype 3 remains a significant cause of morbidity and mortality worldwide, despite inclusion in the 13-valent pneumococcal conjugate vaccine (PCV13). Serotype 3 increased in carriage since the implementation of PCV13 in the USA, while invasive disease rates remain unchanged. We investigated the persistence of serotype 3 in carriage and disease, through genomic analyses of a global sample of 301 serotype 3 isolates of the Netherlands3-31 (PMEN31) clone CC180, combined with associated patient data and PCV utilization among countries of isolate collection. We assessed phenotypic variation between dominant clades in capsule charge (zeta potential), capsular polysaccharide shedding, and susceptibility to opsonophagocytic killing, which have previously been associated with carriage duration, invasiveness, and vaccine escape. We identified a recent shift in the CC180 population attributed to a lineage termed Clade II, which was estimated by Bayesian coalescent analysis to have first appeared in 1968 [95% HPD: 1939-1989] and increased in prevalence and effective population size thereafter. Clade II isolates are divergent from the pre-PCV13 serotype 3 population in non-capsular antigenic composition, competence, and antibiotic susceptibility, the last of which resulting from the acquisition of a Tn916-like conjugative transposon. Differences in recombination rates among clades correlated with variations in the ATP-binding subunit of Clp protease, as well as amino acid substitutions in the comCDE operon. Opsonophagocytic killing assays elucidated the low observed efficacy of PCV13 against serotype 3. Variation in PCV13 use among sampled countries was not independently correlated with the CC180 population shift; therefore, genotypic and phenotypic differences in protein antigens and, in particular, antibiotic resistance may have contributed to the increase of Clade II. Our analysis emphasizes the need for routine, representative sampling of isolates from disperse geographic regions, including historically under-sampled areas. We also highlight the value of genomics in resolving antigenic and epidemiological variations within a serotype, which may have implications for future vaccine development.

Invasive Disease Caused Simultaneously by Dual Serotypes of Streptococcus pneumoniae.

There are at least 98 known pneumococcal serotypes. Invasive pneumococcal disease (IPD) is usually caused by a single serotype, and dual-serotype IPD is rare. To assess factors associated with dual-serotype IPD, patient information obtained through laboratory-based surveillance for IPD from 2005 through 2014 in South Africa was reviewed. Genomes of isolate pairs from coinfected individuals were sequenced to determine their molecular characteristics. For 30 (91%) of 33 patients with dual serotypes, one or both isolates were a pneumococcal conjugate vaccine (PCV13) serotype. Dual-serotype IPD was associated with children <5 years of age (adjusted odds ratio [aOR], 4.7; 95% confidence interval [95% CI], 1.8 to 11.7), underlying illness (other than HIV) (aOR, 2.8; 95% CI, 1.1 to 6.6) and death (aOR, 2.5; 95% CI, 1.08 to 6.09). For each coinfecting pair, isolates were genotypically unrelated, and their genotypes were common among isolates of the same serotype in South Africa. Of 701 accessory genes identified among dual-serotype IPD isolates, four were common between isolate pairs. Coinfecting isolate pairs had different genotypic backgrounds. The association of dual serotypes with death warrants increased awareness of IPD coinfection caused by two or more serotypes.

Global Distribution of Invasive Serotype 35D Streptococcus pneumoniae Isolates following Introduction of 13-Valent Pneumococcal Conjugate Vaccine.

A newly recognized pneumococcal serotype, 35D, which differs from the 35B polysaccharide in structure and serology by not binding to factor serum 35a, was recently reported. The genetic basis for this distinctive serology is due to the presence of an inactivating mutation in wciG, which encodes an O-acetyltransferase responsible for O-acetylation of a galactofuranose. Here, we assessed the genomic data of a worldwide pneumococcal collection to identify serotype 35D isolates and understand their geographical distribution, genetic background, and invasiveness potential. Of 21,980 pneumococcal isolates, 444 were originally typed as serotype 35B by PneumoCaT. Analysis of the wciG gene revealed 23 isolates from carriage (n = 4) and disease (n = 19) with partial or complete loss-of-function mutations, including mutations resulting in premature stop codons (n = 22) and an in-frame mutation (n = 1). These were selected for further analysis. The putative 35D isolates were geographically widespread, and 65.2% (15/23) of them was recovered after the introduction of pneumococcal conjugate vaccine 13 (PCV13). Compared with serotype 35B isolates, putative serotype 35D isolates have higher invasive disease potentials based on odds ratios (OR) (11.58; 95% confidence interval[CI], 1.42 to 94.19 versus 0.61; 95% CI, 0.40 to 0.92) and a higher prevalence of macrolide resistance mediated by mefA (26.1% versus 7.6%; P = 0.009). Using the Quellung reaction, 50% (10/20) of viable isolates were identified as serotype 35D, 25% (5/20) as serotype 35B, and 25% (5/20) as a mixture of 35B/35D. The discrepancy between phenotype and genotype requires further investigation. These findings illustrated a global distribution of an invasive serotype, 35D, among young children post-PCV13 introduction and underlined the invasive potential conferred by the loss of O-acetylation in the pneumococcal capsule.

Phylogenetic Analysis of Invasive Serotype 1 Pneumococcus in South Africa, 1989 to 2013.

Serotype 1 is an important cause of invasive pneumococcal disease in South Africa and has declined following the introduction of the 13-valent pneumococcal conjugate vaccine in 2011. We genetically characterized 912 invasive serotype 1 isolates from 1989 to 2013. Simpson's diversity index (D) and recombination ratios were calculated. Factors associated with sequence types (STs) were assessed. Clonal complex 217 represented 96% (872/912) of the sampled isolates. Following the introduction of the 13-valent pneumococcal conjugate vaccine (PCV13), ST diversity increased in children <5 years (D, 0.39 to 0.63, P = 0.002) and individuals >14 years (D, 0.35 to 0.54, P < 0.001): ST-217 declined proportionately in children <5 years (153/203 [75%] versus 21/37 [57%], P = 0.027) and individuals >14 years (242/305 [79%] versus 96/148 [65%], P = 0.001), whereas ST-9067 increased (4/684 [0.6%] versus 24/228 [11%], P < 0.001). Three subclades were identified within ST-217: ST-217C1 (353/382 [92%]), ST-217C2 (15/382 [4%]), and ST-217C3 (14/382 [4%]). ST-217C2, ST-217C3, and single-locus variant (SLV) ST-8314 (20/912 [2%]) were associated with nonsusceptibility to chloramphenicol, tetracycline, and co-trimoxazole. ST-8314 (20/912 [2%]) was also associated with increased nonsusceptibility to penicillin (P < 0.001). ST-217C3 and newly reported ST-9067 had higher recombination ratios than those of ST-217C1 (4.344 versus 0.091, P < 0.001; and 0.086 versus 0.013, P < 0.001, respectively). Increases in genetic diversity were noted post-PCV13, and lineages associated with antimicrobial nonsusceptibility were identified.

A genome-based survey of invasive pneumococci in Norway over four decades reveals lineage-specific responses to vaccination.

BACKGROUND: Streptococcus pneumoniae is a major cause of mortality globally. The introduction of pneumococcal conjugate vaccines (PCVs) has reduced the incidence of the targeted serotypes significantly, but expansion of non-targeted serotypes, serotype replacement, and incomplete vaccine-targeting contribute to pneumococcal disease in the vaccine era. Here, we characterize the changing population genetic landscape of S. pneumoniae in Norway over a 41-year period (1982-2022). METHODS: Since 2018, all cases of invasive pneumococcal disease have undergone whole-genome sequencing (WGS) at the Norwegian Institute of Public Health. In order to characterize the changing population over time, historical isolates were re-cultured and sequenced, resulting in a historical WGS dataset. Isolates were assigned to global pneumococcal sequence clusters (GPSCs) using PathogenWatch and assigned to serotypes using in silico (SeroBA) and in vitro methods (Quellung reaction). Temporal phylogenetic analyses were performed on GPSCs of particular interest. RESULTS: The availability of WGS data allowed us to study capsular variation at the level of individual lineages. We detect highly divergent fates for different GPSCs following the introduction of PCVs. For two out of eight major GPSCs, we identified multiple instances of serotype switching from vaccine types to non-vaccine types. Dating analyses suggest that most instances of serotype switching predated the introduction of PCVs, but expansion occurred after their introduction. Furthermore, selection for penicillin non-susceptibility was not a driving force for the changing serotype distribution within the GPSCs over time. CONCLUSIONS: PCVs have been major shapers of the Norwegian disease-causing pneumococcal population, both at the level of serotype distributions and the underlying lineage dynamics. Overall, the introduction of PCVs has reduced the incidence of invasive disease. However, some GPSCs initially dominated by vaccine types escaped the effect of vaccination through expansion of non-vaccine serotypes. Close monitoring of circulating lineages and serotypes will be key for ensuring optimal vaccination coverage going forward.

Deep sequencing of Escherichia coli exposes colonisation diversity and impact of antibiotics in Punjab, Pakistan.

Multi-drug resistant (MDR) E. coli constitute a major public health burden globally, reaching the highest prevalence in the global south yet frequently flowing with travellers to other regions. However, our comprehension of the entire genetic diversity of E. coli colonising local populations remains limited. We quantified this diversity, its associated antimicrobial resistance (AMR), and assessed the impact of antibiotic use by recruiting 494 outpatients and 423 community dwellers in the Punjab province, Pakistan. Rectal swab and stool samples were cultured on CLED agar and DNA extracted from plate sweeps was sequenced en masse to capture both the genetic and AMR diversity of E. coli. We assembled 5,247 E. coli genomes from 1,411 samples, displaying marked genetic diversity in gut colonisation. Compared with high income countries, the Punjabi population generally showed a markedly different distribution of genetic lineages and AMR determinants, while use of antibiotics elevated the prevalence of well-known globally circulating MDR clinical strains. These findings implicate that longitudinal multi-regional genomics-based surveillance of both colonisation and infections is a prerequisite for developing mechanistic understanding of the interplay between ecology and evolution in the maintenance and dissemination of (MDR) E. coli.

Detecting co-selection through excess linkage disequilibrium in bacterial genomes.

Population genomics has revolutionized our ability to study bacterial evolution by enabling data-driven discovery of the genetic architecture of trait variation. Genome-wide association studies (GWAS) have more recently become accompanied by genome-wide epistasis and co-selection (GWES) analysis, which offers a phenotype-free approach to generating hypotheses about selective processes that simultaneously impact multiple loci across the genome. However, existing GWES methods only consider associations between distant pairs of loci within the genome due to the strong impact of linkage-disequilibrium (LD) over short distances. Based on the general functional organisation of genomes it is nevertheless expected that majority of co-selection and epistasis will act within relatively short genomic proximity, on co-variation occurring within genes and their promoter regions, and within operons. Here, we introduce LDWeaver, which enables an exhaustive GWES across both short- and long-range LD, to disentangle likely neutral co-variation from selection. We demonstrate the ability of LDWeaver to efficiently generate hypotheses about co-selection using large genomic surveys of multiple major human bacterial pathogen species and validate several findings using functional annotation and phenotypic measurements. Our approach will facilitate the study of bacterial evolution in the light of rapidly expanding population genomic data.

Modulation of multidrug-resistant clone success in Escherichia coli populations: a longitudinal, multi-country, genomic and antibiotic usage cohort study.

BACKGROUND: The effect of antibiotic usage on the success of multidrug-resistant (MDR) clones in a population remains unclear. With this genomics-based molecular epidemiology study, we aimed to investigate the contribution of antibiotic use to Escherichia coli clone success, relative to intra-strain competition for colonisation and infection. METHODS: We sequenced all the available E coli bloodstream infection isolates provided by the British Society for Antimicrobial Chemotherapy (BSAC) from 2012 to 2017 (n=718) and combined these with published data from the UK (2001-11; n=1090) and Norway (2002-17; n=3254). Defined daily dose (DDD) data from the European Centre for Disease Prevention and Control (retrieved on Sept 21, 2021) for major antibiotic classes (ß-lactam, tetracycline, macrolide, sulfonamide, quinolone, and non-penicillin ß-lactam) were used together with sequence typing, resistance profiling, regression analysis, and non-neutral Wright-Fisher simulation-based modelling to enable systematic comparison of resistance levels, clone success, and antibiotic usage between the UK and Norway. FINDINGS: Sequence type (ST)73, ST131, ST95, and ST69 accounted for 892 (49·3%) of 1808 isolates in the BSAC collection. In the UK, the proportion of ST69 increased between 2001-10 and 2011-17 (p=0·0004), whereas the proportions of ST73 and ST95 did not vary between periods. ST131 expanded quickly after its emergence in 2003 and its prevalence remained consistent throughout the study period (apart from a brief decrease in 2009-10). The extended-spectrum ß-lactamase (ESBL)-carrying, globally disseminated MDR clone ST131-C2 showed overall greater success in the UK (154 [56·8%] of 271 isolates in 2003-17) compared with Norway (51 [18·3%] of 278 isolates in 2002-17; p<0·0001). DDD data indicated higher total use of antimicrobials in the UK, driven mainly by the class of non-penicillin ß-lactams, which were used between 2·7-times and 5·1-times more in the UK per annum (ratio mean 3·7 [SD 0·8]). This difference was associated with the higher success of the MDR clone ST131-C2 (pseudo-R2 69·1%). A non-neutral Wright-Fisher model replicated the observed expansion of non-MDR and MDR sequence types under higher DDD regimes. INTERPRETATION: Our study indicates that resistance profiles of contemporaneously successful clones can vary substantially, warranting caution in the interpretation of correlations between aggregate measures of resistance and antibiotic usage. Our study further suggests that in countries with low-to-moderate use of antibiotics, such as the UK and Norway, the extent of non-penicillin ß-lactam use modulates rather than determines the success of widely disseminated MDR ESBL-carrying E coli clones. Detailed understanding of underlying causal drivers of success is important for improved control of resistant pathogens. FUNDING: Trond Mohn Foundation, Marie Sklodowska-Curie Actions, European Research Council, Royal Society, and Wellcome Trust.

Impact of pneumococcal conjugate vaccines on invasive pneumococcal disease-causing lineages among South African children.

Invasive pneumococcal disease (IPD) due to non-vaccine serotypes after the introduction of pneumococcal conjugate vaccines (PCV) remains a global concern. This study used pathogen genomics to evaluate changes in invasive pneumococcal lineages before, during and after vaccine introduction in South Africa. We included genomes (N = 3104) of IPD isolates from individuals aged <18 years (2005-20), spanning four periods: pre-PCV, PCV7, early-PCV13, and late-PCV13. Significant incidence reductions occurred among vaccine-type lineages in the late-PCV13 period compared to the pre-PCV period. However, some vaccine-type lineages continued to cause invasive disease and showed increasing effective population size trends in the post-PCV era. A significant increase in lineage diversity was observed from the PCV7 period to the early-PCV13 period (Simpson's diversity index: 0.954, 95% confidence interval 0.948-0.961 vs 0.965, 0.962-0.969) supporting intervention-driven population structure perturbation. Increases in the prevalence of penicillin, erythromycin, and multidrug resistance were observed among non-vaccine serotypes in the late-PCV13 period compared to the pre-PCV period. In this work we highlight the importance of continued genomic surveillance to monitor disease-causing lineages post vaccination to support policy-making and future vaccine designs and considerations.

Pan-pathogen deep sequencing of nosocomial bacterial pathogens in Italy in spring 2020: a prospective cohort study.

BACKGROUND: Nosocomial infections pose a considerable risk to patients who are susceptible, and this is particularly acute in intensive care units when hospital-associated bacteria are endemic. During the first wave of the COVID-19 pandemic, the surge of patients presented a significant obstacle to the effectiveness of infection control measures. We aimed to assess the risks and extent of nosocomial pathogen transmission under a high patient burden by designing a novel bacterial pan-pathogen deep-sequencing approach that could be integrated with standard clinical surveillance and diagnostics workflows. METHODS: We did a prospective cohort study in a region of northern Italy that was severely affected by the first wave of the COVID-19 pandemic. Inpatients on both ordinary and intensive care unit (ICU) wards at the San Matteo hospital, Pavia were sampled on multiple occasions to identify bacterial pathogens from respiratory, nasal, and rectal samples. Diagnostic samples collected between April 7 and May 10, 2020 were cultured on six different selective media designed to enrich for Acinetobacter baumannii, Escherichia coli, Enterococcus faecium, Enterococcus faecalis, Klebsiella pneumoniae, Pseudomonas aeruginosa, Staphylococcus aureus, and Streptococcus pneumoniae, and DNA from each plate with positive growth was deep sequenced en masse. We used mSWEEP and mGEMS to bin sequencing reads by sequence cluster for each species, followed by mapping with snippy to generate high quality alignments. Antimicrobial resistance genes were detected by use of ARIBA and CARD. Estimates of hospital transmission were obtained from pairwise bacterial single nucleotide polymorphism distances, partitioned by within-patient and between-patient samples. Finally, we compared the accuracy of our binned Acinetobacter baumannii genomes with those obtained by single colony whole-genome sequencing of isolates from the same hospital. FINDINGS: We recruited patients from March 1 to May 7, 2020. The pathogen population among the patients was large and diverse, with 2148 species detections overall among the 2418 sequenced samples from the 256 patients. In total, 55 sequence clusters from key pathogen species were detected at least five times. The antimicrobial resistance gene prevalence was correspondingly high, with key carbapenemase and extended spectrum ß-lactamase genes detected in at least 50 (40%) of 125 patients in ICUs. Using high-resolution mapping to infer transmission, we established that hospital transmission was likely to be a significant mode of acquisition for each of the pathogen species. Finally, comparison with single colony Acinetobacter baumannii genomes showed that the resolution offered by deep sequencing was equivalent to single-colony sequencing, with the additional benefit of detection of co-colonisation of highly similar strains. INTERPRETATION: Our study shows that a culture-based deep-sequencing approach is a possible route towards improving future pathogen surveillance and infection control at hospitals. Future studies should be designed to directly compare the accuracy, cost, and feasibility of culture-based deep sequencing with single colony whole-genome sequencing on a range of bacterial species. FUNDING: Wellcome Trust, European Research Council, Academy of Finland Flagship program, Trond Mohn Foundation, and Research Council of Norway.

Inferring the heritability of bacterial traits in the era of machine learning.

Quantification of heritability is a fundamental desideratum in genetics, which allows an assessment of the contribution of additive genetic variation to the variability of a trait of interest. The traditional computational approaches for assessing the heritability of a trait have been developed in the field of quantitative genetics. However, the rise of modern population genomics with large sample sizes has led to the development of several new machine learning-based approaches to inferring heritability. In this article, we systematically summarize recent advances in machine learning which can be used to infer heritability. We focus on an application of these methods to bacterial genomes, where heritability plays a key role in understanding phenotypes such as antibiotic resistance and virulence, which are particularly important due to the rising frequency of antimicrobial resistance. By designing a heritability model incorporating realistic patterns of genome-wide linkage disequilibrium for a frequently recombining bacterial pathogen, we test the performance of a wide spectrum of different inference methods, including also GCTA. In addition to the synthetic data benchmark, we present a comparison of the methods for antibiotic resistance traits for multiple bacterial pathogens. Insights from the benchmarking and real data analyses indicate a highly variable performance of the different methods and suggest that heritability inference would likely benefit from tailoring of the methods to the specific genetic architecture of the target organism. Availability and implementation: The R codes and data used in the numerical experiments are available at: https://github.com/tienmt/her_MLs.

Mge-cluster: a reference-free approach for typing bacterial plasmids.

Extrachromosomal elements of bacterial cells such as plasmids are notorious for their importance in evolution and adaptation to changing ecology. However, high-resolution population-wide analysis of plasmids has only become accessible recently with the advent of scalable long-read sequencing technology. Current typing methods for the classification of plasmids remain limited in their scope which motivated us to develop a computationally efficient approach to simultaneously recognize novel types and classify plasmids into previously identified groups. Here, we introduce mge-cluster that can easily handle thousands of input sequences which are compressed using a unitig representation in a de Bruijn graph. Our approach offers a faster runtime than existing algorithms, with moderate memory usage, and enables an intuitive visualization, classification and clustering scheme that users can explore interactively within a single framework. Mge-cluster platform for plasmid analysis can be easily distributed and replicated, enabling a consistent labelling of plasmids across past, present, and future sequence collections. We underscore the advantages of our approach by analysing a population-wide plasmid data set obtained from the opportunistic pathogen Escherichia coli, studying the prevalence of the colistin resistance gene mcr-1.1 within the plasmid population, and describing an instance of resistance plasmid transmission within a hospital environment.

Evolutionary and functional history of the Escherichia coli K1 capsule.

Escherichia coli is a leading cause of invasive bacterial infections in humans. Capsule polysaccharide has an important role in bacterial pathogenesis, and the K1 capsule has been firmly established as one of the most potent capsule types in E. coli through its association with severe infections. However, little is known about its distribution, evolution and functions across the E. coli phylogeny, which is fundamental to elucidating its role in the expansion of successful lineages. Using systematic surveys of invasive E. coli isolates, we show that the K1-cps locus is present in a quarter of bloodstream infection isolates and has emerged in at least four different extraintestinal pathogenic E. coli (ExPEC) phylogroups independently in the last 500 years. Phenotypic assessment demonstrates that K1 capsule synthesis enhances E. coli survival in human serum independent of genetic background, and that therapeutic targeting of the K1 capsule re-sensitizes E. coli from distinct genetic backgrounds to human serum. Our study highlights that assessing the evolutionary and functional properties of bacterial virulence factors at population levels is important to better monitor and predict the emergence of virulent clones, and to also inform therapies and preventive medicine to effectively control bacterial infections whilst significantly lowering antibiotic usage.

Key features of pneumococcal isolates recovered in Central and Northwestern Russia in 2011-2018 determined through whole-genome sequencing.

Invasive pneumococcal disease remains one of the leading causes of morbidity and mortality worldwide. In Russia, 13- valent pneumococcal conjugate vaccine (PCV13) was introduced into the childhood immunization programme nationwide in 2014. As part of the Global Pneumococcal Sequencing Project (GPS), we used genome data to characterize 179 pneumococcal isolates collected from Russia in 2011-2018 to investigate the circulating pneumococcal strains using a standardized genomic definition of pneumococcal lineages (global pneumococcal sequence clusters, GPSCs), prevalent serotypes and antimicrobial resistance profiles.We observed high serotype and lineage diversity among the 179 isolates recovered from cerebrospinal fluid (n=77), nasopharyngeal swabs (n=99) and other non-sterile site swabs (n=3). Overall, 60 GPSCs were identified, including 48 clonal complexes (CCs) and 14 singletons, and expressed 42 serotypes (including non-typable). Among PCV13 serotypes, 19F, 6B and 23F were the top three serotypes while 11A, 15B/C and 8 were the top three among non-PCV13 serotypes in the collection. Two lineages (GPSC6 and GPSC47) expressed both PCV13 and non-PCV13 serotypes that caused invasive disease, and were penicillin- and multidrug-resistant (MDR), highlighting their potential to adapt and continue to cause infections under vaccine and antibiotic selective pressure. PCV13 serotypes comprised 92 % (11/12) of the CSF isolates from the children aged below 5 years; however, the prevalence of PCV13 serotype isolates dropped to 53 % (31/58) among the nasopharyngeal isolates. Our analysis showed that 59 % (105/179) of the isolates were predicted to be non-susceptible to at least one class of antibiotics and 26 % (46/179) were MDR. Four MDR lineages (GPSC1, GPSC6, GPSC10 and GPSC47) accounted for 65 % (30/46) of the MDR isolates and expressed PCV13 serotypes (93 %, 28/30).This study provides evidence of high genetic and serotype diversity contributed by a mix of globally spreading and regionally circulating lineages in Russia. The observations suggest that the PCV13 vaccine could be important in reducing both invasive disease and antimicrobial resistance. We also identify potential lineages (GPSC6 and GPSC47) that may evade the vaccine.

Changes in serotype prevalence of Streptococcus pneumoniae in Southampton, UK between 2006 and 2018.

Streptococcus pneumoniae continues to cause significant disease burden. Whilst pneumococcal conjugate vaccines (PCV) have substantially reduced this burden, serotype replacement partially negates this success due to increased disease associated with non-vaccine serotypes (NVTs). Continued surveillance is therefore essential to provide crucial epidemiological data. Annual cross-sectional surveillance of paediatric pneumococcal carriage was started in Southampton, UK following PCV7 roll-out in 2006. Nasopharyngeal swabs were collected from children < 5 years old each winter (October to March) from 2006/07 and for each consecutive year until 2017/18. Pneumococcal serotype was inferred from whole genome sequencing data. A total of 1429 (32.5%) pneumococci were isolated from 4093 children. Carriage ranged from 27.8% (95%CI 23.7-32.7) in 2008/09 to 37.9% (95%CI 32.8-43.2) in 2014/15. Analyses showed that carriage increased in children aged 24-35 months (p < 0.001) and 47-60 months (p < 0.05). Carriage of PCV serotypes decreased markedly following PCV7 and/or PCV13 introduction, apart from serotype 3 where the relative frequency was slightly lower post-PCV13 (pre-PCV13 n = 7, 1.67%; post-PCV13 n = 13, 1.27%). Prevalence of NVTs implicated in increased disease was low with 24F (n = 19, 1.4%) being the most common followed by 9N (n = 11, 0.8%), 8 (n = 7, 0.5%) and 12F (n = 3, 0.2%).